Microgrid energy intelligent management system and method based on Internet of Things

Abstract

The invention discloses a micro-grid energy intelligent management system and method based on the Internet of Things, and relates to the technical field of power grid management. Historical power utilization data of electric appliances in a micro-grid are collected, and a power utilization power prediction function is established; collecting power generation historical data of the generator sets, establishing power generation power prediction functions, when a certain generator set fails, setting the generator set as a target generator set and recording a fault moment, extracting parameter change characteristics and comparing the parameter change characteristics with a fault classification model, determining a target fault, collecting and superposing the power generation power prediction functions of all the generator sets except the target generator set, and calculating the power generation power of the generator sets. The method comprises the steps of obtaining a power generation prediction target function, obtaining a power protection value corresponding to a target fault, obtaining a power utilization power prediction function and an average power generation power reduction amount calculation risk value under the target fault, and achieving hierarchical management and control from fault recognition to influence evaluation to active early warning through layer-by-layer progression of a classification model, a quantitative index and the prediction function. The system reliability is improved.

Description

Technical Field

[0001] This invention relates to the field of power grid management technology, specifically to a microgrid energy intelligent management system and method based on the Internet of Things. Background Technology

[0002] As a new type of power system integrating renewable energy generation, intelligent power distribution, and load management, the stability and reliability of microgrids are crucial for the efficient utilization of distributed energy resources. Microgrids typically include various types of generators such as photovoltaic and wind power, operate in complex environments, and experience diverse fault types, with different faults having significantly different impacts on power generation.

[0003] Therefore, traditional microgrid fault management methods have the following shortcomings: 1. Traditional microgrid management methods mainly rely on electrical parameters of grid equipment or human experience for fault identification, lacking integrated analysis of multi-source parameters such as electrical, mechanical, and environmental factors, making it difficult to accurately capture the dynamic characteristics of faults; 2. The quantitative analysis of power loss and repair time caused by faults is not refined according to fault type, making it impossible to make risk judgments for the coupling of various types of generator units in the microgrid system, and to provide differentiated basis for resource scheduling and reserve capacity configuration; 3. There is a lack of dynamic risk models that couple multiple variables such as fault characteristics, unit attributes, and power loss, ignoring the dynamic changing characteristics of fault impacts. Summary of the Invention

[0004] The purpose of this invention is to provide a microgrid energy intelligent management system and method based on the Internet of Things, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a microgrid energy intelligent management method based on the Internet of Things, the method comprising: Step S100: Obtain historical maintenance records and historical operation records of generator sets in the microgrid, and classify the faults of generator sets according to their fault characteristics; Step S200: Collect its historical maintenance records, count the repair time of various faults and calculate the average value. At the time of each fault occurrence, extract the average power generation within the time period before and after that point, calculate the reduction in power generation before and after the fault, and classify and count the average reduction by fault type. Step S300: After a microgrid generator set fails, obtain the historical records of all generator set failures within a unit sampling period, calculate the probability of occurrence of each type of failure in each unit time period, obtain the average power reduction and probability of occurrence for each type of failure, and, in conjunction with the total number of generator sets, calculate the power protection value for each unit time period and establish the correspondence between the power protection value and the type of failure. Step S400: Collect historical electricity consumption data of electrical appliances in the microgrid, obtain the characteristics of power consumption change over time, establish a power consumption prediction function that predicts power consumption over time, collect historical power generation data of generator sets in the microgrid, obtain the characteristics of power generation change over time, and establish a power generation prediction function for each generator set. Step S500: When a generator set in the microgrid fails, the generator set is set as the target unit and the time of failure is recorded as the target time. The historical operation records of the target unit are collected, the parameter change characteristics are extracted and compared with the fault classification model. The fault type with the highest similarity is taken as the target fault. The power generation prediction functions of all units in the microgrid except the target unit in this time period are collected and superimposed to obtain the power generation prediction target function of the microgrid in the target time period. Step S600: Obtain the power protection value for each time period corresponding to the target fault, and at the same time obtain the power consumption prediction function of the microgrid appliances and the average power reduction of the target generator set under the target fault within the time period. Calculate the risk value for each time period within the target time period. When the risk value is less than the threshold, issue an alarm to the microgrid management personnel.

[0006] Furthermore, step S100 includes: Step S101: Obtain the time point when the generator set malfunctioned from the generator set's operation log, set a unit time period, and obtain the historical operation records of the generator set in the unit time period before and after the time point. Step S102: Obtain the change characteristics of generator set operating parameters from historical operation records, use the change characteristics of operating parameters as classification labels, and classify the classification labels through machine learning. Each classification category corresponds to a fault type. The generator operating parameters include electrical parameters, mechanical parameters and environmental parameters. The generator operating parameters also include the reduction in power generation after the generator set has a fault. The change characteristics include the time domain characteristics and frequency domain characteristics of the generator operating parameters. Step S103: Collect the correspondence between each fault type and the changing characteristics of generator set operating parameters to obtain a fault type classification model.

[0007] By collecting historical maintenance and operation records, generator set faults are classified based on fault characteristics, and a systematic fault labeling system is established. This avoids the limitations of single-dimensional fault analysis. By comprehensively collecting multi-source data, including electrical parameters such as voltage and current, mechanical parameters such as temperature and cooling system, and environmental parameters such as vibration and noise, fault characteristics are fully captured, laying the foundation for subsequent refined analysis.

[0008] Furthermore, step S200 includes: Step S201: Collect historical maintenance records of the same generator set, obtain the fault repair time of the generator set after each fault, classify the fault repair time according to the fault type, and calculate the average fault repair time for each fault type. Step S202: Obtain the time point when the generator set fails each time, collect the average power generation in the unit time period before the time point and the average power generation in the unit time period after the time point, and calculate the reduction in power generation of the generator set before and after the failure. Step S203: Classify the reduction in power generation according to the type of fault, and calculate the average reduction in power generation for each type of fault.

[0009] The average repair time is calculated based on the type of fault, and the average reduction in power generation for different fault types is calculated to quantify the sustained and intensive impacts of the faults.

[0010] Furthermore, step S300 includes: Step S301: When any generator set in the microgrid fails, k unit time periods are combined into a sampling period. The failure records of all generator sets in the microgrid are obtained in a sampling period after the failure occurs. The probability of occurrence of each failure type in each unit time period is calculated. Step S302: In the m-th unit time period of a sampling period, obtain the average power reduction p of the i-th fault type. im The probability η of the i-th fault type occurring in the microgrid during the m-th unit time period. i Calculate the power protection value H for the m-th unit time period. m , Where C represents the total number of fault types, R i This represents the number of all generator sets in the microgrid whose historical fault records include the i-th fault type, excluding those that have already failed. Step S303: Compile the power protection values ​​into a success rate protection value sequence according to the order of the unit time period, and establish a correspondence between the fault type and the power protection value sequence.

[0011] The power protection value is calculated by multiplying the fault probability, the number of units, and the power loss within a unit time period, forming a protection value sequence. This transforms fault risk into a quantifiable power protection indicator, reflecting the power fluctuation risk that the system may face in different time periods, and dynamically adjusting the protection strategy to avoid being overly conservative or insufficient.

[0012] Furthermore, step S500 includes: Step S501: When a generator set in the microgrid fails, set the generator set as the target generator set, obtain the time when the target generator set fails, and record the time as the target time. Step S502: Collect the historical operation records of the target generator set in the unit time period before and after the target time, extract the change characteristics of the target generator set's operating parameters, compare them with the fault classification model, and select the fault type with the highest similarity as the target fault type. Step S503: Take the average fault repair time corresponding to the target fault, and the time period after the target time as the target time period. Collect the power generation prediction functions of all generator sets in the microgrid except the target generator set in the target time period, and superimpose them to obtain the power generation prediction objective function G of the microgrid in the target time period.

[0013] Shorten fault diagnosis time and avoid the inefficiency of manual troubleshooting; dynamically define the impact period based on the repair time, accurately assess the system's power generation capacity during the fault, and provide real-time data support for subsequent risk calculations.

[0014] Furthermore, step S600 includes: Step S601: Obtain the power protection value sequence corresponding to the target fault, and obtain the power protection value for each unit time period in the target time period; Step S602: Obtain the power consumption prediction function of electrical appliances in the microgrid during the target time period, denoted as the power consumption prediction target function F, and obtain the average power generation reduction q of the target generator set when the target fault type occurs; Step S603: Calculate the risk value M of the i-th unit time period in the target time period. i M i =(D(ti)-F(ti)) min -H i -q, where D(ti) represents the predicted power generation of all generators in the microgrid excluding the target generator set at time ti in the i-th unit time period of the target time period, and F(ti) represents the predicted power consumption of electrical appliances in the microgrid at time ti in the i-th unit time period of the target time period. min Let Hi represent the minimum value function, Hi represent the power protection value of the i-th unit time period in the target time period, and q represent the reduction in the power generation of the target generator set; Step S604: When M i When the value is less than 0, the i-th unit time period in the target time period is recorded as the unit risk time period. All unit risk time periods in the target time period are collected and an alarm is issued to the microgrid management personnel.

[0015] It enables multi-dimensional coupled analysis of "supply and demand gap - potential risks - failure losses", timely alerts for insufficient power, avoids power outages, and the alarm mechanism allows managers to call backup power in advance, improving system robustness.

[0016] To better implement the above methods, an IoT-based microgrid energy intelligent management system is also proposed. The system includes: a fault classification module, a fault power management module, a fault propagation management module, a power prediction module, a target time period management module, and a risk management module. The system includes the following modules: Fault Classification Module (for collecting fault characteristics of generator sets and classifying faults); Fault Power Management Module (for managing power generation losses caused to the microgrid after generator set faults); Fault Propagation Management Module (for managing the propagation records of generator set faults in the microgrid); Power Prediction Module (for collecting power generation and consumption records of the microgrid and establishing power generation and consumption prediction functions); Target Time Period Management Module (for obtaining the power generation prediction target function for the target time period in the current fault of the microgrid); and Risk Management Module (for collecting prediction information of the microgrid and alerting management personnel when risks exist in the microgrid).

[0017] Furthermore, the fault classification module includes: a historical record management unit, a feature extraction unit, and a classification model management unit. The historical record management unit is used to manage the historical operating records of generator sets in the microgrid, the feature extraction unit is used to extract the changing features in the historical operating parameters of the generator sets, and the classification model management unit is used to manage the classification model of fault types.

[0018] Furthermore, the fault power management module includes a repair time management unit and a power loss management unit. The repair time management unit is used to manage the repair time after a generator set fails, and the power loss management unit is used to manage the amount of power generation loss of the generator after a failure.

[0019] Furthermore, the fault propagation management module includes: a diffusion rate management unit, a power protection value calculation unit, and a protection value sequence management unit. The diffusion rate management unit is used to manage the failure probability of other generator sets after a failure of a generator set in the microgrid. The power protection value calculation unit is used to calculate the power protection value for each unit time period. The protection value sequence management unit is used to manage the correspondence between fault types and power protection value sequences.

[0020] Furthermore, the target time period management module includes: a fault information management unit, a fault feature comparison unit, and a target time period management unit. The fault information management unit is used to obtain fault information of generator sets in the microgrid. The fault feature comparison unit is used to extract the fault features of the current generator set, compare them with the classification model, and manage the fault types of the target faults. The target time period management unit is used to manage the target time period according to the correspondence between fault types and average fault repair time.

[0021] Furthermore, the risk management module includes: a predicted value management unit, a risk value management unit, and an alarm unit. The predicted value management unit is used to manage the predicted power consumption and power generation values ​​within the target time period. The risk value management unit is used to calculate the risk value within the target time period. The alarm unit is used to aggregate the time periods within the target time period that meet the alarm conditions and to provide alarm prompts to the management personnel.

[0022] Compared with existing technologies, the beneficial effects of this invention are: historical data collection and real-time risk warning form a data closed loop, supporting the digitalization and intelligentization of microgrid fault management; through the progressive development of classification models, quantitative indicators, and prediction functions, it achieves hierarchical control from fault identification to impact assessment and then to proactive warning, improving system reliability; it supports differentiated analysis of multiple types of units, constructs a cross-equipment risk assessment system, is applicable to complex microgrid environments, and provides technical assurance for the stability of multi-energy complementary systems. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the structure of a microgrid energy intelligent management system based on the Internet of Things according to the present invention; Figure 2 This is a schematic diagram of the structure of a microgrid energy intelligent management method based on the Internet of Things according to the present invention. Detailed Implementation

[0024] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] Example: Figures 1-2 As shown, the present invention provides a technical solution, a microgrid energy intelligent management method based on the Internet of Things, the method comprising: Step S100: Obtain historical maintenance records and historical operation records of generator sets in the microgrid, and classify the faults of generator sets according to their fault characteristics; Step S100 includes: Step S101: Obtain the time point when the generator set malfunctioned from the generator set's operation log, set a unit time period, and obtain the historical operation records of the generator set in the unit time period before and after the time point. Step S102: Obtain the change characteristics of generator set operating parameters from historical operation records, use the change characteristics of operating parameters as classification labels, and classify the classification labels through machine learning. Each classification category corresponds to a fault type. The generator operating parameters include electrical parameters, mechanical parameters and environmental parameters. The generator operating parameters also include the reduction in power generation after the generator set has a fault. The change characteristics include the time domain characteristics and frequency domain characteristics of the generator operating parameters. Step S103: Collect the correspondence between each fault type and the changing characteristics of generator set operating parameters to obtain a fault type classification model.

[0026] In this embodiment, each tag type can be encoded, and the tags can be clustered. Each cluster of a category tag is considered as a fault type. The clustering core of the fault types is used as the clustering feature. The clustering features of all fault types are aggregated. By aggregating the operating parameters of the generator set under different random risks, the impact of random risks on the generator set can be reflected. In this embodiment, the electrical parameters include the generator set's voltage parameters, current parameters, electrical frequency parameters, and electrical phase parameters; the mechanical parameters include equipment temperature and cooling system parameters; and the environmental parameters include vibration parameters and noise levels.

[0027] Step S200: Collect its historical maintenance records, count the repair time of various faults and calculate the average value. At the time of each fault occurrence, extract the average power generation within the time period before and after that point, calculate the reduction in power generation before and after the fault, and classify and count the average reduction by fault type. Step S200 includes: Step S201: Collect historical maintenance records of the same generator set, obtain the fault repair time of the generator set after each fault, classify the fault repair time according to the fault type, and calculate the average fault repair time for each fault type. Step S202: Obtain the time point when the generator set fails each time, collect the average power generation in the unit time period before the time point and the average power generation in the unit time period after the time point, and calculate the reduction in power generation of the generator set before and after the failure. Step S203: Classify the reduction in power generation according to the type of fault, and calculate the average reduction in power generation for each type of fault.

[0028] In the embodiments, generator sets can be classified according to their type, and the failure probability of different types of generator sets can be statistically analyzed. For example, in a microgrid system, there are 20 photovoltaic generator sets and 5 wind turbine generator sets. When a certain type of short-term strong wind disaster occurs, if the support frame of a photovoltaic generator unit breaks and the photovoltaic panel collapses, the power generation of the photovoltaic panel will drop from 550W to 0, and the power generation will decrease by 550W. Wind turbine generators activate self-protection mechanisms, such as actively adjusting the blade pitch angle to gradually deviate from the wind direction, reducing wind energy capture, lowering the generator's speed and power output, and preventing equipment overload operation. When affected by one of the aforementioned short-term strong wind disasters, the wind turbine will activate its protection mechanism, reducing its power generation from 300kW to 100kW, a reduction of 200kW.

[0029] Step S300: After a microgrid generator set fails, obtain the historical records of all generator set failures within a unit sampling period, calculate the probability of occurrence of each type of failure in each unit time period, obtain the average power reduction and probability of occurrence for each type of failure, and, in conjunction with the total number of generator sets, calculate the power protection value for each unit time period and establish the correspondence between the power protection value and the type of failure. Step S300 includes: Step S301: When any generator set in the microgrid fails, k unit time periods are combined into a sampling period. The failure records of all generator sets in the microgrid are obtained in a sampling period after the failure occurs. The probability of occurrence of each failure type in each unit time period is calculated. Step S302: In the m-th unit time period of a sampling period, obtain the average power reduction pim of the i-th fault type, the probability ηi of the i-th fault type occurring in the microgrid in the m-th unit time period, and calculate the power protection value H for the m-th unit time period. m , Where C represents the total number of fault types, and Ri represents the number of all generator sets in the microgrid whose historical fault records include the i-th fault type, excluding the generator sets that have already failed. Step S303: Compile the power protection values ​​into a success rate protection value sequence according to the order of the unit time period, and establish a correspondence between the fault type and the power protection value sequence.

[0030] In the first time period, the probability that the photovoltaic panels of the remaining photovoltaic generator sets will collapse, causing the generator set's power output to drop from 550W to 0, is 0.4. The probability that the wind turbine generator set's protection mechanism will activate, causing the power output to drop from 300kW to 100kW, is 0.15. In the second time period, the probability that the photovoltaic panels of the remaining photovoltaic generator sets will collapse, causing the generator set's power output to drop from 550W to 0, is 0.45. The probability that the wind turbine generator set's protection mechanism will activate, causing the power output to drop from 300kW to 100kW, is 0.25. In the third time period, the probability that the photovoltaic panels of the remaining photovoltaic generator sets will collapse, causing the generator set's power output to drop from 550W to 0, is 0.1. The probability that the wind turbine generator set's protection mechanism will activate, causing the power output of a single wind turbine generator to drop from 300kW to 100kW, is 0.05. In the second time period, the power reduction due to the first type of fault (photovoltaic generator failure) is 19 × 0.4 × 550 = 4180W, and the power reduction due to the second type of fault (wind turbine failure) is 5 × 0.15 × 200kW = 150kW. The power protection value for the first time period is 154.18kW. In the second time period, the power reduction of the first type of fault, namely the photovoltaic generator set fault, is 19×0.45×550=4703W, and the power reduction of the second type of fault, namely the wind turbine generator set fault, is 5×0.1×200kW=100kW. The power protection value in the second time period is approximately 104.7kW. In the third time period, the power reduction of the first type of fault, namely the photovoltaic generator set fault, is 19×0.1×550=1045W, and the power reduction of the second type of fault, namely the wind turbine generator set fault, is 5×0.05×200kW=50kW. The power protection value in the third time period is approximately 51kW.

[0031] Therefore, under a certain operating parameter classification, the success rate protection value sequence when the photovoltaic generator power output drops from 550W to 0 is 154.18kW, 104.7kW, and 51kW.

[0032] Step S400: Collect historical electricity consumption data of electrical appliances in the microgrid, obtain the characteristics of power consumption change over time, establish a power consumption prediction function that predicts power consumption change over time, collect historical power generation data of generator sets in the microgrid, obtain the characteristics of power generation change over time, and establish a power generation prediction function for each generator set.

[0033] In this embodiment, historical data on power generation and consumption are converted into time series, and a prediction function for power generation to power consumption is constructed using an LSTM model.

[0034] Step S500: When a generator set in the microgrid fails, the generator set is set as the target unit and the time of failure is recorded as the target time. The historical operation records of the target unit are collected, the parameter change characteristics are extracted and compared with the fault classification model. The fault type with the highest similarity is taken as the target fault. The power generation prediction functions of all units in the microgrid except the target unit in this time period are collected and superimposed to obtain the power generation prediction target function of the microgrid in the target time period. Step S500 includes: Step S501: When a generator set in the microgrid fails, set the generator set as the target generator set, obtain the time when the target generator set fails, and record the time as the target time. Step S502: Collect the historical operation records of the target generator set in the unit time period before and after the target time, extract the change characteristics of the target generator set's operating parameters, compare them with the fault classification model, and select the fault type with the highest similarity as the target fault type. Step S503: Take the average fault repair time corresponding to the target fault, and the time period after the target time as the target time period. Collect the power generation prediction functions of all generator sets in the microgrid except the target generator set in the target time period, and superimpose them to obtain the power generation prediction objective function G of the microgrid in the target time period.

[0035] Step S600: Obtain the power protection value for each time period corresponding to the target fault, and at the same time obtain the power consumption prediction function of the microgrid appliances and the average power reduction of the target generator set under the target fault within the time period. Calculate the risk value for each time period within the target time period. When the risk value is less than the threshold, issue an alarm to the microgrid management personnel. Step S600 includes: Step S601: Obtain the power protection value sequence corresponding to the target fault, and obtain the power protection value for each unit time period in the target time period; Step S602: Obtain the power consumption prediction function of electrical appliances in the microgrid during the target time period, denoted as the power consumption prediction target function F, and obtain the average power generation reduction q of the target generator set when the target fault type occurs; Step S603: Calculate the risk value Mi of the i-th unit time period in the target time period, Mi = (D(ti) - F(ti)) min -Hi-q, where D(ti) represents the predicted power generation of all generator sets in the microgrid except the target generator set at time ti in the i-th unit time period of the target time period, and F(ti) represents the predicted power consumption of electrical appliances in the microgrid at time ti in the i-th unit time period of the target time period. minLet Hi represent the minimum value function, Hi represent the power protection value of the i-th unit time period in the target time period, and q represent the reduction in the power generation of the target generator set; Step S604: When Mi < 0, the i-th unit time period in the target time period is recorded as the unit risk time period. All unit risk time periods in the target time period are collected and an alarm is sent to the microgrid management personnel.

[0036] In this embodiment, the repair time for the target unit needs to go through 3 unit time periods, of which the first unit time period is (D(t1)-F(t1)). min =190.45kW, in the second unit time period (D(t2)-F(t2)) min =50.25kW, in the 3rd unit time period (D(t3)-F(t3)) min =260.45kW; Calculate M1 = 190.45 - 154.18 - 0.55 = 35.73 > 0, M2 = 50.25 - 104.7 - 0.55 = -55 < 0, and M3 = 260.45 - 51.55 - 0.55 = 208.35 > 0 respectively; After a target unit fails, the system calculates that there may be insufficient power in the microgrid in the second time period after the target unit fails, and reminds the management personnel to call in supplementary power in time.

[0037] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A microgrid energy intelligent management method based on the Internet of Things, characterized in that: The methods include: Step S100: Obtain historical maintenance and operation records of microgrid generator sets, and classify faults according to fault characteristics; Step S200: Collect historical maintenance records, statistically analyze the repair time of various faults and calculate the average value, extract the average power generation of the time period before and after the fault occurrence point, calculate the reduction in power generation before and after the fault and statistically analyze the average reduction by fault type. Step S300: Obtain the historical records of all generator set faults within a unit sampling period after the generator set fault, calculate the probability of occurrence of each fault type in each unit time period, obtain the average power reduction and occurrence probability, and, in combination with the total number of generator sets, calculate the power protection value in a unit time period and establish the correspondence with the fault type. Step S400: Collect historical electricity consumption data of electrical appliances in the microgrid and establish an electricity consumption prediction function; collect historical power generation data of generator sets and establish a power generation prediction function; Step S500: When a generator set fails, it is set as the target unit and the time of failure is recorded. The historical operation record of the target unit is collected, the parameter change characteristics are extracted and compared with the fault classification model to determine the target fault. The power generation prediction functions of all units except the target unit are collected and superimposed to obtain the power generation prediction target function. Step S600: Obtain the power protection value corresponding to the target fault, obtain the power consumption prediction function and calculate the risk value based on the average power reduction under the target fault, and issue an alarm to the management personnel when the risk value is less than the threshold.

2. The microgrid energy intelligent management method based on the Internet of Things according to claim 1, characterized in that: Step S100 includes: Step S101: Obtain the time point when the generator set malfunctioned from the generator set's operation log, set a unit time period, and obtain the historical operation records of the generator set for the unit time period before and after the time point. Step S102: Obtain the change characteristics of generator set operating parameters from historical operation records, use the change characteristics of operating parameters as classification labels, and classify the classification labels through machine learning. Each classification category corresponds to a fault type. The generator operating parameters include electrical parameters, mechanical parameters and environmental parameters. The generator operating parameters also include the reduction in power generation after the generator set has a fault. The change characteristics include the time domain characteristics and frequency domain characteristics of the generator operating parameters. Step S103: Collect the correspondence between each fault type and the changing characteristics of generator set operating parameters to obtain a fault type classification model.

3. The microgrid energy intelligent management method based on the Internet of Things according to claim 2, characterized in that: Step S200 includes: Step S201: Collect historical maintenance records of the same generator set, obtain the fault repair time of the generator set after each fault, classify the fault repair time according to the fault type, and calculate the average fault repair time for each fault type. Step S202: Obtain the time point when the generator set fails each time, collect the average power generation in the unit time period before the time point and the average power generation in the unit time period after the time point, and calculate the reduction in power generation of the generator set before and after the failure. Step S203: Classify the reduction in power generation according to the type of fault, and calculate the average reduction in power generation for each type of fault.

4. The microgrid energy intelligent management method based on the Internet of Things according to claim 3, characterized in that: Step S300 includes: Step S301: When any generator set in the microgrid fails, k unit time periods are combined into a sampling period. The failure records of all generator sets in the microgrid are obtained in a sampling period after the failure occurs. The probability of occurrence of each failure type in each unit time period is calculated. Step S302: In the m-th unit time period of a sampling period, obtain the average power reduction p of the i-th fault type. im The probability η of the i-th fault type occurring in the microgrid during the m-th unit time period. i Calculate the power protection value H for the m-th unit time period. m , Where C represents the total number of fault types, R i This represents the number of all generator sets in the microgrid whose historical fault records include the i-th fault type, excluding those that have already failed. Step S303: Compile the power protection values ​​into a success rate protection value sequence according to the order of the unit time period, and establish a correspondence between the fault type and the power protection value sequence.

5. The microgrid energy intelligent management method based on the Internet of Things according to claim 4, characterized in that: Step S500 includes: Step S501: When a generator set in the microgrid fails, the generator set is set as the target generator set, the time when the target generator set fails is obtained, and the time is recorded as the target time. Step S502: Collect the historical operation records of the target generator set in the unit time period before and after the target time, extract the change characteristics of the target generator set's operating parameters, compare them with the fault classification model, and select the fault type with the highest similarity as the target fault type. Step S503: Take the average fault repair time corresponding to the target fault, and the time period after the target time as the target time period. Collect the power generation prediction functions of all generator sets in the microgrid except the target generator set in the target time period, and superimpose them to obtain the power generation prediction objective function G of the microgrid in the target time period.

6. The microgrid energy intelligent management method based on the Internet of Things according to claim 5, characterized in that: Step S600 includes: Step S601: Obtain the power protection value sequence corresponding to the target fault, and obtain the power protection value for each unit time period in the target time period; Step S602: Obtain the power consumption prediction function of electrical appliances in the microgrid during the target time period, denoted as the power consumption prediction target function F, and obtain the average power generation reduction q of the target generator set when the target fault type occurs; Step S603: Calculate the risk value M of the i-th unit time period in the target time period. i M i =(D(ti)-F(ti)) min -H i -q, where D(ti) represents the predicted power generation of all generators in the microgrid excluding the target generator set at time ti in the i-th unit time period of the target time period, and F(ti) represents the predicted power consumption of electrical appliances in the microgrid at time ti in the i-th unit time period of the target time period. min Let Hi represent the minimum value function, Hi represent the power protection value of the i-th unit time period in the target time period, and q represent the reduction in the power generation of the target generator set; Step S604: When M i When the value is less than 0, the i-th unit time period in the target time period is recorded as the unit risk time period. All unit risk time periods in the target time period are collected and an alarm is issued to the microgrid management personnel.

7. A microgrid energy intelligent management system based on the Internet of Things (IoT), used to execute the microgrid energy intelligent management method based on the Internet of Things as described in any one of claims 1-6, characterized in that: The system includes: Fault classification module, fault power management module, fault propagation management module, power prediction module, target time period management module, and risk management module; The system includes the following modules: Fault Classification Module (for collecting fault characteristics of generator sets and classifying faults); Fault Power Management Module (for managing power generation losses caused to the microgrid after generator set faults); Fault Propagation Management Module (for managing the propagation records of generator set faults in the microgrid); Power Prediction Module (for collecting power generation and consumption records of the microgrid and establishing power generation and consumption prediction functions); Target Time Period Management Module (for obtaining the power generation prediction target function for the target time period in the current fault of the microgrid); and Risk Management Module (for collecting prediction information of the microgrid and alerting management personnel when risks exist in the microgrid).

8. The microgrid energy intelligent management system based on the Internet of Things according to claim 7, characterized in that: The fault classification module includes: a historical record management unit, a feature extraction unit, and a classification model management unit. The historical record management unit is used to manage the historical operation records of generator sets in the microgrid, the feature extraction unit is used to extract the changing features in the historical operation parameters of the generator sets, and the classification model management unit is used to manage the classification model of fault types. The fault power management module includes a repair time management unit and a power loss management unit. The repair time management unit is used to manage the repair time after a generator set fails, and the power loss management unit is used to manage the amount of power loss of the generator after a failure.

9. A microgrid energy intelligent management system based on the Internet of Things according to claim 7, characterized in that: The fault propagation management module includes: a diffusion rate management unit, a power protection value calculation unit, and a protection value sequence management unit. The diffusion rate management unit is used to manage the failure probability of other generator sets after a failure of a generator set in the microgrid. The power protection value calculation unit is used to calculate the power protection value for each unit time period. The protection value sequence management unit is used to manage the correspondence between fault types and power protection value sequences.

10. A microgrid energy intelligent management system based on the Internet of Things according to claim 7, characterized in that: The target time period management module includes: a fault information management unit, a fault feature comparison unit, and a target time period management unit. The fault information management unit is used to obtain fault information of generator sets in the microgrid. The fault feature comparison unit is used to extract the fault features of the current generator set, compare them with the classification model, and manage the types of target faults. The target time period management unit is used to manage the target time period according to the correspondence between fault types and average fault repair time. The risk management module includes a forecast value management unit, a risk value management unit, and an alarm unit. The forecast value management unit is used to manage the forecast values ​​of power consumption and power generation in the target time period. The risk value management unit is used to calculate the risk value in the target time period. The alarm unit is used to collect the unit time periods that meet the alarm conditions in the target time period and provide alarm prompts to the management personnel.

Images